Inhibitors of HDACs modulate transcription and induce cell growth arrest, differentiation, and apoptosis. HDAC inhibitors (HDACIs) also enhance the cytotoxic effects of therapeutic agents used in cancer treatment, including radiation and chemotherapeutic drugs. Moreover, recent research indicates that transcriptional dysregulation may contribute to the molecular pathogenesis of certain neurodegenerative disorders, such as Huntington's disease, Rett syndrome, Charcot-Marie-Tooth disease (CMT) and other peripheral neuropathies, spinal muscular atrophy, amyotropic lateral sclerosis, and ischemia. For example, suberoylanilide hydroxamic acid (SAHA) has been shown to penetrate into the brain to dramatically improve motor impairment in a mouse model of Huntington's disease, thereby validating research directed to HDACIs in the treatment of neurodegenerative diseases. Furthermore, selective HDAC6 inhibitors have been shown to rescue the CMT phenotype, restore proper mitochondrial motility, and correct the axonal transport defects observed in transgenic mice. Selective HDAC6 inhibitors also induce the re-innervation of muscles and increase the number of observed neuromuscular junctions in these same models (C. d'Ydewalle et al., Nature Medicine 2011).
A recent review summarized evidence that aberrant histone acetyltransferase (HAT) and HDAC activity may be a common underlying mechanism contributing to neurodegeneration. Moreover, from a mouse model of depression, the therapeutic potential of HDACs in treating depression is discussed. See WO 2008/019025, designating the United States, incorporated herein in its entirety.
Eleven isozymes in the HDAC family of enzymes, which can be grouped into classes by their evolutionary relationships, have been identified. Structure and function appear to be conserved among members of the various classes. The HDAC family is made up of class I HDACs, including HDAC1, 2, 3, and 8; class IIa, including HDAC4, 5, 7, and 9; class IIb, including HDAC6 and 10; and a class IV enzyme, HDAC11 (A. J. de Ruijter et al., The Biochemical Journal 2003, 370(Pt), 737-749).
The class I HDACs are found primarily in the nucleus and are expressed in all tissue types, except for the muscle cell-specific HDAC8. The class I HDACs interact with many key transcription factors regulating gene expression, including CoREST and NuRD. Class IIa HDACs have tissue specific expression, and are found in both the nucleus and cytoplasm. Unlike the other isozymes, the class IIb HDAC6 does not extensively associate with transcription factors, and acts as a deacetylase on non-histone proteins, including α-tubulin, HSP90, cortactin, and the peroxiredoxins (O. Witt et al., Cancer Letters 2008; R. B. Parmigiana et al., PNAS 2008).
HDACs form multiprotein complexes with many regulatory proteins inside the cell. For example, HDAC4, 5, and 7 actually lack intrinsic deacetylase ability, and gain activity only by interacting with HDAC3. Each isozyme interacts with a specific series of regulatory proteins and transcription factors and has a specific set of substrates, and thus each regulates a specific series of genes and proteins (O. Witt et al., Cancer Letters 2008). The design of selective HDAC isozyme inhibitors allows preferential inhibition of only the isozyme(s) relevant to a particular disease or condition, thereby reducing the probability of counterproductive and/or adverse effects resulting from an unwanted and undesired inhibition of other HDAC isozymes.
HDAC6 is the most abundant histone deacetylase isozyme in the human body, and along with HDAC7, is the most commonly expressed isozyme in the brain (A. J. de Ruijter et al., The Biochemical Journal 2003, 370(Pt), 737-749). Structurally significant features of HDAC6 include two deacetylase domains and a zinc finger motif. It is most commonly found in the cytoplasm, but can be shuttled into the nucleus via its nuclear export signal. A cytoplasmic retention signal, which sequesters the enzyme in the cytoplasm, also was found (A. Valenzuela-Fernandez et al., Trends in Cell Biology 2008, 18(6), 291-297). The functions of HDAC6 are unlike any of the other HDAC isozymes. Many non-histone substrates are deacetylated by HDAC6, including α-tubulin, HSP90, cortactin, and peroxiredoxins (O. Witt et al., Cancer Letters 2008; R. B. Parmigiani et al., PNAS USA 2008, 105 (28), 9633-9638).
The design of HDACIs focuses on the three major domains of the enzyme molecule. A zinc binding group (ZBG) of the HDACI typically is a hydroxamic acid, benzamide, or thiol, although other functional groups have been used. This ZBG moiety of the inhibitor chelates the zinc cofactor found in the active site of the enzyme. The ZBG moiety typically is bonded to a lipophilic linker group, which occupies a narrow channel leading from the HDAC surface to the active site. This linker, in turn, is bonded to a surface recognition, or ‘cap’, moiety, which typically is an aromatic group that interacts with residues at the surface of the enzyme (K. V. Butler et al., Current Pharmaceutical Design 2008, 14(6), 505-528).
Consideration of each structural element is important in the design of HDACIs (37). Alteration of the ZBG has profound effects on inhibitor potency. The most potent inhibitors frequently feature a hydroxamic acid ZBG, though other groups such as ketones, amides, and thiols effectively inhibit the enzyme with lower potency. Low molecular weight compounds having carboxylic acid ZBGs, such as valproic acid, inhibit HDACs at micromolar potency, but have profound effects in vivo when given in high doses. Hydroxamic acids chelate zinc in a bidentate fashion and hydrogen bond with H142 and H143 (HDAC8), as revealed by crystal structure data. Computational studies have refined this description of chelation, demonstrating that the hydroxamic acid carbonyl coordinates zinc more strongly than the hydroxyl group. Energetically favorable interactions between hydroxamic acid and the active site provide for high potency inhibition, but the hydroxamic acid functional group has some undesirable properties from a drug design perspective. Hydroxamic acids are potentially mutagenic, and have given positive results in the Ames test. Hydroxamic acids also are promiscuous zinc chelators and potently inhibit many zinc-containing metalloproteins. The replacement of hydroxamic acid with a different potent ZBG is an active area of research in HDACI design and development.
The linker region typically is a hydrophobic aryl or alkyl scaffold, which occupies the hydrophobic HDAC catalytic channel. Most reported HDACIs, including SAHA, feature an alkyl chain linker, mimicking the lysine alkyl chain. Aromatic groups are frequently included in the linker region of an HDACI, for example, in the HDACIs panbinostat and belinostat.
Manipulation of the cap group can greatly increase potency because this group has the potential to interact with multiple residues at the enzyme surface. The most potent inhibitors typically feature an aryl cap group scaffold. Biaryl and heteroaromatic cap group scaffolds have been extensively investigated. The large tetrapeptide motif of apicidin imparts high potency, and gives high potency with a number of diverse ZBGs. The tetrapeptide cap group motif is common to natural product HDACIs, and has been engineered to produce libraries of tetrapeptide HDACIs for use in screening protocols.
Currently, at least eleven HDACIs are in clinical development. These HDACIs can be divided into at least five chemical classes, illustrated below, based on their structure, and in most cases they broadly and nonselectively inhibit class I/II HDACs with varying efficiency. These five chemical classes are hydroxamates, cyclic tetrapeptides, cyclic peptides, short-chain fatty acids, and benzamides. Typically, known HDACIs fail to show prominent HDAC isozyme selectivity, which as stated above can cause serious problems in a clinical setting, especially in the treatment of diseases and conditions wherein a prolonged drug administration of an HDACI is required. For example, it has been found that some HDACIs enhance lung and microglial inflammation (TSA and SAHA), as well as high glucose-induced inflammation. If this effect is linked to specific HDAC isozymes, the use of certain HDACIs would be contraindicated in various diseases and conditions, such as diabetes and asthma.

Additional HDACI's include

The following table summarizes some HDACIs that presently are in clinical trials.
TABLE IInhibitorIndicationsSAHAT-cell lymphoma (Approved)RomidepsinT-cell lymphoma (Approved)Multiple myeloma (Phase III)Peripheral T-cell lymphoma (Phase III)Refractory renal cell cancer (Phase II)Valproic AcidBipolar disorder (Approved)Acute myeloid leukemia (Phase I/II with all trans-retinoic acid)PCI-24781Leukemia (Phase I/II)ITF-2357Hodgkins lymphoma (Phase II)Follicular lymphoma (Phase III, with yttrium-90-ibritumomab)Juvenile arthritis (Phase II)Myeloproliferative Diseases (Phase II)MS-275MelanomaLymphoma (halted due to dose limiting toxicities)Advanced acute leukemias (Phase 1)Combination trials with DNA methyltransferaseinhibitors and 5-azacitidine in non-small cell lung cancer(Preclinical)PanbinostatT-cell lymphoma (Phase II)Prostate cancer (Phase I with docetaxel)BelinostatSolid tumors (Phase I)Mesothelioma (Abandoned)MGCD0103Solid tumors (Phase II with gemcitabine)Diffuse large B-cell lymphoma (Phase II)EVP-0334Parkinson's disease (Phase I)
Clinical trial information relating to HDACIs is published in J. Tan et al., Journal of hematology & oncology. 3:5 (2010) and L. Wang et al., Nat Rev Drug Discov. 8:969-81 (2009).
HDAC-regulated factors have been implicated in the mechanisms of major central nervous system (CNS) disorders. In Parkinson's disease (PD), α-synuclein binds to histones and inhibits HAT activity, causing neurodegeneration. Application of HDACIs to PD neurons blocks α-synuclein toxicity. Dysregulation of histone acetylation, involving CBP, a neuroprotective transcription factor with histone acetyltransferase activity, has been found in Huntington's disease (HD), Alzheimer's disease (AD), and Rubinstein-Taybi syndrome (T. Abel et al., Curr. Opin. in Pharmacol. 2008, 8(1), 57-64). In a cellular model of AD, cell death was accompanied by loss of CBP function and histone deacetylation. The mutant HD protein, htt, interacts with CBP, inhibiting the HAT activity and causing cell death. Treatment with an HDACI helps to restore histone acetylation, protecting against neurodegeneration and improving motor performance in a mouse model of HD (C. Rouaux et al., Biochem. Pharmacol. 2004, 68(6), 1157-1164).
Various studies directed to the application of HDACIs in the context of CNS disorders have implicated the class II HDACs, particularly HDAC6, as potential therapeutic targets. One investigation revealed that inhibition of HDAC6 could be beneficial as a treatment for HD, a disease for which no pharmacological treatment is available. The mutant htt protein found in HD disrupts intracellular transport of the pro-survival and pro-growth nerve factor, BDNF, along the microtubule network, causing neuronal toxicity. Inhibition of HDAC6 promotes transport of BDNF by promoting tubulin hyperacetylation. TSA (trichostatin A), a nonselective HDAC inhibitor, was found to facilitate transport and release of BNDF-containing vesicles (J. P. Dompierre et al., J Neurosci 2007, 27(13), 3571-3583). These results provide a biological basis for the identification and development of HDACIs, and particularly HDAC6 selective inhibitors, as a treatment for HD and other neurodegenerative disorders.
HDACIs prevent or delay neuronal dysfunction and death in in vitro and in vivo models thereby indicating that HDACIs are broadly neuroprotective. For example, HDACIs have shown therapeutic efficacy in the polyglutamine-expansion disorder Huntington's disease. While the neuroprotective mechanisms of the HDACIs in rodent models are not yet understood, it is clear that HDACIs induce the expression of certain genes that confer neuroprotection. The upregulation of HSP-70 and Bcl-2 through the inhibition of HDAC has been observed in the cortex and striatum of rats after focal cerebral ischemia. HSP-70 expression has been found to result in neuroprotection in a number of disease models including Alzheimer's disease (AD), Parkinson's disease (PD), and Huntington's disease (HD). In addition, HDAC6 inhibition leads to the acetylation of peroxiredoxin and increases its antioxidant activity which may contribute to the neuroprotective effects of these compounds (R. B. Parmigiana et al., PNAS 2008).
Studies also provide good evidence that HDACI-induced p21cip1/waf1 expression may play a significant role in HDACI-mediated neuroprotection. It recently was reported that p21cip1/waf1 overexpression protects neurons from oxidative stress-induced death, that p21cip1/waf1 is induced in the rodent brain by HDAC inhibition, and that homozygous loss of p21cip1/waf1 exacerbates damage in a mouse MCAO/reperfusion model of ischemic stroke. In a similar study, the HDAC inhibitor TSA was shown to increase gelsolin expression in neurons, and that gelsolin expression is necessary for neuroprotection in an oxygen/glucose deprivation model of neurodegeneration and a mouse MCAO/reperfusion model of ischemic stroke.
Alternatively, unrelated to histone acetylation and gene upregulation, proteins such as α-tubulin and HSP90 are targets for acetylation and become acetylated when HDACs are inhibited. In tumor cells, the acetylation of HSP90 has been shown to decrease the ability of HSP90 to interact with certain client proteins and thereby abrogate chaperone function. With regard to stroke and traumatic brain injury (TBI), as well as several other neurodegenerative diseases, the inhibition of HSP90 is predicted to have a positive effect on neuronal survival. Indeed, the pharmacological HSP90 inhibitor, Geldanamycin, and its analogs have been shown to be neuroprotective in a number of stroke models. HSP90 inhibition and the consequent release of heat-shock factor (HSF) to the nucleus may also, in part, explain an upregulation of HSP70 in the brain during focal ischemia and HDACI treatment.
In addition, HDACIs are useful in the treatment of cancers. For example, histone acetylation and deacetylation play important roles in chromatin folding and maintenance (Kornberg et al., Bjorklund et al., Cell, 1999, 96:759-767; Struhl et al., Cell, 1998, 94:1-4). Acetylated chromatin is more open and has been implicated in the increased radiation sensitivities observed in some cell types (Oleinick et al., Int. J. Radiat. Biol. 1994, 66:523-529). Furthermore, certain radiation-resistant human cancer cells treated with the HDACI inhibitor TSA were sensitized to the damaging effects of ionizing radiation. Thus, HDACIs appear useful as radiation sensitizing agents.
WO 2008/055068, designating the U.S. and incorporated herein in its entirety, discloses numerous diseases and conditions treatable by HDACIs, including the underlying science and reasoning supporting such treatments.
HDAC6 therefore has emerged as an attractive target for drug development and research. (C. M. Grozinger et al., Proc. Natl. Acad. Sci. USA 1999, 96, 4868-73; and C. Boyault et al., Oncogene 2007, 26, 5468-76.) Presently, HDAC6 inhibition is believed to offer potential therapies for autoimmunity, cancer, and many neurodegenerative conditions. (S. Minucci et al., Nat. Rev. Cancer. 2006, 6, 38-51; L. Wang et al., Nat. Rev. Drug Discov. 2009, 8, 969-81; J. P. Dompierre et al., J. Neurosci. 2007, 27, 3571-83; and A. G. Kazantsev et al., Nat. Rev. Drug Discov. 2008, 7, 854-68.) Selective inhibition of HDAC6 by small molecule or genetic tools has been demonstrated to promote survival and re-growth of neurons following injury, offering the possibility for pharmacological intervention in both CNS injury and neurodegenerative conditions. (M. A. Rivieccio et al., Proc. Natl. Acad. Sci. USA 2009, 106, 19599-604.) Unlike other histone deacetylases, inhibition of HDAC6 does not appear to be associated with any toxicity, making it an excellent drug target. (O. Witt et al., Cancer Lett 2009, 277, 8-21.) Tubacin, an HDAC6 selective inhibitor, used in models of disease, has helped to validate, in part, HDAC6 as a drug target, but its non-drug-like structure, high lipophilicity (ClogP=6.36 (KOWWIN)) and tedious synthesis make it more useful as a research tool than a drug. (S. Haggarty et al., Proc. Natl. Acad. Sci. USA 2003, 100, 4389-94.) Other compounds also have a modest preference for inhibiting HDAC6. (S. Schafer et al., ChemMedChem 2009, 4, 283-90; Y. Itoh et al., J. Med. Chem. 2007, 50, 5425-38; and S. Manku et al., Bioorg. Med. Chem. Lett. 2009, 19, 1866-70.)
In summary, extensive evidence supports a therapeutic role for HDACIs in the treatment of a variety of conditions and diseases, such as cancers and CNS diseases and degenerations. However, despite exhibiting overall beneficial effects, like beneficial neuroprotective effects, for example, HDACIs known to date have little specificity with regard to HDAC inhibition, and therefore inhibit all zinc-dependent histone deacetylases. It is still unknown which is (are) the salient HDAC(s) that mediate(s) neuroprotection when inhibited. Emerging evidence suggests that at least some of the HDAC isozymes are absolutely required for the maintenance and survival of neurons, e.g., HDAC1. Additionally, adverse side effect issues have been noted with nonspecific HDAC inhibition. Thus, the clinical efficacy of present-day nonspecific HDACIs for stroke, neurodegenerative disorders, neurological diseases, and other diseases and conditions ultimately may be limited. It is important therefore to design, synthesize, and test compounds capable of serving as potent, and preferably isozyme-selective, HDACIs that are able to ameliorate the effects of neurological disease, neurodegenerative disorder, traumatic brain injury, cancer, inflammation, malaria, autoimmune diseases, immunosuppressive therapy, and other conditions and diseases mediated by HDACs.
An important advance in the art would be the discovery of HDACIs, and particularly selective HDAC6 inhibitors, that are useful in the treatment of diseases wherein HDAC inhibition provides a benefit, such as cancers, neurological diseases, traumatic brain injury, neurodegenerative disorders and other peripheral neuropathies, stroke, hypertension, malaria, allograft rejection, rheumatoid arthritis, and inflammations. Accordingly, a significant need exists in the art for efficacious compounds, compositions, and methods useful in the treatment of such diseases, alone or in conjunction with other therapies used to treat these diseases and conditions. The present invention is directed to meeting this need.